Abstract:

As Intelligent Tutoring Systems evolve to become an essential medium for educational content delivery, a challenge presents itself concerning the development of high-quality educational content that meets the needs of its audience. The development of such content is a process that requires many resources and expertise.

In this research, we propose an innovative system aimed at transforming educational content creation by making use of Natural Language Processing (NLP) and Large Language Models (LLMs). Particularly using Generative Pre-Trained Transformers (GPT). The approach taken in the design of this system makes use of prompt engineering, a process where the inputted prompt is tailored and refined to lead the LLM into providing a favourable result, as a result, we can generate content that is aligned with the needs of the material’s audience.

Having carried out empirical testing of the system with educators, the evaluation of our system indicates favourable results due to the high acceptance rate present in the feedback obtained. Approximately 84% of the system’s users expressed their satisfaction with the content generated, while 82% expressed that the content generated was indeed accurate and 83% of the users indicated that they would be willing to make use of this content in their classroom.

Despite the clear benefits such research entails, ethical concerns did not go uninvestigated. When making use of any system that processes sensitive data, which an educational system is bound to do, it is important to have the appropriate measures in place such that users' rights to data privacy are not violated, on this end of the spectrum the system does not retain or use the inputted data beyond generating the appropriate output.

This research aims to further explore the limits of automated content creation. Future work will focus on expanding the abilities and applicability of this system as well as the investigation of its potential impact on the learning process.

Keywords:

NLP, technology, education, personalisation, automated content creation.